Skip Navigation

the Environmental and Energy Resources Library

Home Browse Resources Get Recommendations Forums About Help Advanced Search

Shared and Distributed Memory Parallelization of a Lagrangian Atmospheric Dispersion Model

This paper describes parallelization of a 3-D Lagrangian stochastic atmospheric dispersion model using both distributed- and shared-memory methods. Shared-memory parallelism is implemented through the use of OpenMP compiler directives. Distributed-memory parallelism relies on the MPI message-passing library. One or both of the parallel modes can be used depending upon the requirements of the problem and the computational platform available. The distributed-memory version achieves a nearly linear decrease in execution time as the number of processors is increased. As the number of particles per processor is lowered, performance is limited by the decrease in work per processor and by the need to produce one set of output files. The shared memory version achieves a speed-up factor of approximately 1.4 running on machines with four processors.

Cumulative Rating: (not yet rated)
Date Of Record Release 2009-06-15 15:14:02
Description This paper describes parallelization of a 3-D Lagrangian stochastic atmospheric dispersion model using both distributed- and shared-memory methods. Shared-memory parallelism is implemented through the use of OpenMP compiler directives. Distributed-memory parallelism relies on the MPI message-passing library. One or both of the parallel modes can be used depending upon the requirements of the problem and the computational platform available. The distributed-memory version achieves a nearly linear decrease in execution time as the number of processors is increased. As the number of particles per processor is lowered, performance is limited by the decrease in work per processor and by the need to produce one set of output files. The shared memory version achieves a speed-up factor of approximately 1.4 running on machines with four processors.
Classification
Resource Type
Format
Subject
Source National Atmospheric Release Advisory Center
Keyword 3-D, Parallelization, Stochastic atmospheric dispersion model, Emergency planning
Selector Stith
Date Of Record Creation 2009-06-15 15:05:53
Education Level
Date Last Modified 2010-04-09 17:10:20
Creator David J. Larson, J. S. Nasstrom
Language English
Date Record Checked: 2009-06-15 00:00:00 (W3C-DTF)

Log In:





Mission
EERL's mission is to be the best possible online collection of environmental and energy sustainability resources for community college educators and for their students. The resources are also available for practitioners and the public.

EERL & ATEEC
EERL is a product of a community college-based National Science Foundation Center, the Advanced Technology Environmental and Energy Center (ATEEC), and its partners.

Contact ATEEC 563.441.4087 or by email ateec@eicc.edu